Effect of Target Change During the Simple Attack in Fencing

Journal of Sports Sciences, 2013 Vol. 31, No. 10, 1100–1107, http://dx.doi.org/10.1080/02640414.2013.770908

Effect of target change during the simple attack in fencing

MARCOS GUTIÉRREZ-DÁVILA1, F. JAVIER ROJAS1, MATTEO CALETTI2, RAQUEL ANTONIO3, & ENRIQUE NAVARRO3

1University of Granada, Faculty of Sport Sciences, Department of Physical Education and Sports, Granada, Spain, 2University of Granada, Faculty of Psychology, Granada, Spain, and 3Technical University of Madrid, Faculty of Physical Activity and Sport Sciences, Madrid, Spain

(Accepted 9 January 2013)

Abstract The aim of this study was to test the effect that changing targets during a simple long lunge attack in fencing exerts on the temporal parameters of the reaction response, the execution speed, and the precision and the coordination of the movement pattern. Thirty fencers with more than 10 years of experience participated in this study. Two force platforms were used to record the horizontal components of the reaction forces and thereby to determine the beginning of the movement. A three- dimensional (3D) system recorded the spatial positions of the 9 markers situated on the fencer plus the epee, while a moving target was projected on a screen, enabling the control of the target change. The results indicated that when a target change is provoked the reaction time (RT), movement time (MT), and the time used in the acceleration phase of the centre of mass (CM) increases signiﬁcantly with respect to the attack executed with a straight thrust. The speed and horizontal distance reached by the CM at the end of the acceleration phase (VX(CM) and SX(CM), respectively) signiﬁcantly decreased, while the errors increased. However, the temporal sequence of the movement pattern did not appreciably change.

Keywords: motor control, biomechanics, fencer, reaction response

Introduction must wait until the opponent starts the defensive action (Movement Time (MT)-stimulus change). The lunge is one of the most frequently used actions Then, when the defensive action is detected (S ), during the attack in fencing. This technical action is 2 the attacker should inhibit the initial attack, process executed in an uninterrupted way with a temporal the information, and make the most appropriate sequence that begins with the acceleration of the decision while accelerating forwards, this being con- epee arm, then the back foot becomes active fol- sidered “choice reaction time” (CRT). Finally, the lowed by the front foot, and ends when the point of fencer needs sufﬁcient movement time to change the the epee touches the target. During the movement, a aim of the epee and reach the target (MT-target horizontal thrust phase can be distinguished, when

Downloaded by [UGR-BTCA Gral Universitaria] at 13:15 14 May 2013 change). Figure 1 shows a sequence of a simple the centre of mass (CM) of the fencer accelerates attack together with the movement phases (upper forwards, leading to a flight phase that ends when the part), as well as a simplified time model of the simple front foot makes contact with the floor (Stewart & attack with only a direct thrust (A) together with a Kopetka, 2005). The movement pattern is based on simple attack involving a target change (B). a model of pushing kinetic chain (Kreighbaum & In the practice of fencing, the simple attack with Barthels, 1996), which facilitates the change in the target change, described in Figure 1 (B), is usually trajectory of the weapon at any instant during the trained for individually, with the coach (master) act- action. ing as opponent and attempting to simulate a real Following classical fencing manuals, the lunge can action (Czajkowski, 2005). Recently, automatic sys- be executed by a straight thrust or by disengage- tems that permit a simulation of this situation ment, changing the target during the action through randomised control of stimuli (Electronic (Thirioux, 1970). Sometimes, the disengagement is Training Target, Favero Electronic Design - Italy) used to change the target during the attack in have been used. Besides the acquisition of a certain response to a defence action. In this case, the fencer functional variability in movement patterns, this type

Correspondence: F. Javier Rojas, University of Granada, Faculty of Sport Sciences, Department of Physical Education and Sports, Granada, Spain. Email: [email protected]

Figure 1. Schematic representation of the simple attack sequence together with the movement phases (upper part), as well as a simpliﬁed temporal model of the simple attack executed with a direct thrust (A) and of the simple attack with a target change (B). (RT = reaction time.)

of training is intended to help the fencer reduce their In another sphere, cognitive neuropsychologists choice reaction time, attempting to associate every recognise that the motor planes and their sequencing response with a stimulus. However, studies carried arise in an implicit, unconscious, and automatic way, out that entail varying levels of uncertainty of whereas the intentions of the movements are con- response for the fencer (Borysiuk & Waskiewicz, scious and can be influenced by prior information, 2008; Sanderson, 1983), along with certain explana- giving rise to simple facilitation or priming, which tory theories discussed below, cause us to suspect favours rapid responses (Desmurget & Sirigu, 2009). that the uncertainty produced by target change dur- However, to guarantee adaptability in the responses ing the execution of a movement may reduce the when a target change is required during an action, a velocity of displacement and modify the coordina- facilitation or high-order strategy is necessary to tion of the movement pattern, approximating the enable the inhibition of the first action triggered by context of real competition. a stimulus to refine the precision of the response From the dual-visual-system model, the reaction (Gao, Wong-Lin, Holmes, Simen, & Cohen, 2009). Downloaded by [UGR-BTCA Gral Universitaria] at 13:15 14 May 2013 response requires the contribution of the two visual According to Duque, Lew, Mazzocchio, Oliver, processes that have different functions. A ventral and Ivry (2010), this inhibition process develops by current, associated with explicit awareness, provides two closely related mechanisms. The first inhibits information on the probabilities to execute the action the activation of possible responses selected at the (affordable perception), while a dorsal current gath- spinal level (impulse control to avoid premature ers visual cues implicit in movement control, imme- reactions) and the second determines one response diately and relatively quickly (Goodale & Westwood, among the most relevant options (conflict resolu- 2004; Milner & Goodale, 1995). Following Van der tion). Thus, before the movement starts, all the pos- Kamp, Rivas, Van Doorn, and Savelsbergh (2008), sible responses are activated, requiring inhibitory the ventral current would be more involved before signals at the spinal level, anticipating external infor- starting the movement, while the dorsal would dom- mation that would cause one response to prevail over inate over the ventral during movement execution. all others. This second inhibitory mechanism would Accordingly, in the case of a target change during a occur at the upper cortical levels, causing a certain simple attack, the dominance of the dorsal current delay in the response (Ivanoff, Branning & Marois, would be delayed until the instant at which the target 2009; Schluter, Rushworth, Passingham, & Mills, change occurs, thus slowing down the initial 1998). This aspect could be related to the findings movement. of Di Russo, Taddei, Apnile, and Spinelli (2006), 1102 M. Gutiérrez-Dávila et al.

who compared high-level fencers with non-athlete the forte of the epee blade (near the hilt). A projector university students, and showed that the capacity connected to a computer with a programmable that fencers have to change one target to a more external card enabled the timed projection of a circle appropriate one is the result of greater inhibitory of 0.09 m over a white surface of 0.70 × 0.55 m activity by the prefrontal cortex over motor representing a plastron, after a random period of processes. 0.5 s to 1.2 s, from the starting cue. After this period, According to the above-mentioned researchers, an electronic signal was used to synchronise all the the target change in response to the opponent’s recording systems mentioned. defensive action increases uncertainty, delays the Following the protocol used by Williams and dominance of the dorsal current during the attack, Walmsley (2000), for a long lunge the plastron was and activates the inhibitory mechanism of conflict situated at a distance of 1.5-fold the height of the resolution before spurring movement. This research fencer, with respect to the big toe of the back foot. suggests the hypothesis that, in the situation of sim- After several simple attacks with a straight thrust at ple attack with target change, there will be an the pre-established distance, the plastron was moved increase in the two temporal components of the to a distance at which the fencer felt comfortable. reaction response time (RRT): a) simple reaction After the projection of the circle was set to the estab- time (RT) and b) movement time (MT) as conse- lished diameter, the distance chosen by each partici- quence of the reduction in horizontal velocity of the pant was maintained for all the trials in the two centre of mass (CM). experimental situations (straight thrust and target change). After the adjustment, the fencers reduced the distance to the plastron by 0.013 ± 0.2 m. The Methods fencers received instructions to remain still in the on-guard position until the circle was projected on Participants the plastron. From this instant, they were instructed A total of 30 fencers participated, each with experi- to make a simple attack with a straight thrust as ence in national competition for more than 10 years. rapidly as possible, touching the point of the epee Of these, 13 were members of the National Spanish within the circle. Before recordings began, the fen- Fencing Team and the other 17 were participants of cers made several thrusts until becoming completely national ranking in Spain (age = 35.2 ± 10.4 years familiar with the measurement system. In order to old; height = 1.79 ± 0.05 m; body mass = 82 ± 10 kg). maintain the level of activation of the participants, All procedures were conducted according to the they were informed of the times obtained after every ethical guidelines of the university. All fencers were trial. To confirm the effect exerted by target change thoroughly informed about the study and their par- in fencing, we compared two types of simple attack ticipation, and gave their written consent prior to (direct thrust and target change). commencement. Direct-thrust attack. In the first recording phase, the fencers had to respond with a simple attack of a Procedures straight long lunge when the circle appeared, stimu- The fencers used their personal epee equipped with lus 1 (S1), in the centre of the plastron. Five valid an electronic chronometer with a precision of 1 ms trials were made for all the participants in which the Downloaded by [UGR-BTCA Gral Universitaria] at 13:15 14 May 2013 (0.001 s), adapted to a system that recorded the time reaction response time (RRT) was recorded, the needed in each trial for the point of the epee to touch errors being recorded when the point of the epee the plastron. The fencers adopted their customary did not reach the circle. Following the method of on-guard position, placing their feet on the two force Gutiérrez-Dávila, Dapena, and Campos (2006), the platforms (0.6 × 0.37 m), Dinascan/IBV (IBV, start of the movement was determined from the Valencia, Spain), which operated at 500 Hz, instant at which the net force of the horizontal com- enabling us to record the horizontal component ponent (FAX +FBX) reached a value greater than or of the reaction force (FAX and FBX). A three- equal to 1% of body weight. When this time was less dimensional (3D) movement-recording system than 100 ms, the trial was repeated. VICON-460, (Vicon Motion Systems, Oxford, UK), with 6 infrared video cameras, functioning at Target change attack. After the recording of 5 valid 500 Hz recorded the spatial positions of 8 body trials corresponding to the first phase, the tests were markers situated on the two heels, the points of the made for the simple attack with a target change, fi shoes, the knee (epicondylus bularis femoris), the stimuli 2 (S2). As in the previous situation, begin- hip (trochanter major), the shoulder (tuberculum ning from the on-guard position, the fencer lunged majus) and the elbow (epicondylus radialis) of the as swiftly as possible, seeking to touch the point of epee arm, plus a marker situated at the beginning of the epee within the circle that appeared in the centre Effect of target change in fencing 1103

of the plastron. The participants were aware from the of the movements of the centre of mass (CM) of the beginning of the second phase that the target might fencer system plus the epee, the velocity and distance move, but they did not know what the nature of of the horizontal component travelled by the CM those movements might be. The target could move during the first 0.2 s of the acceleration phase were to three equidistant positions situated on both sides measured (VX(CM) and SX(CM) respectively). For the of and below the starting position (circle situated at analysis of movement, we have used a system of the centre of the plastron) at a distance of 0.25 m inertial references associated with the ground, from the initial target. where the horizontal axis (X) has been identified To determine the instant at which the position of with the principal direction of movement (Figure 1). the circle would change (time-S2), the time compo- With the aim of determining the possible changes nents of the median reaction response time of the 5 in the starting position (on-guard), the positions valid trials of the simple attack with a straight thrust were recorded by the markers situated on the hip, were used (time-S2 = Reaction Time + ¼Movement shoulder and epee, with respect to the position of the Time; see Figure 1). S2 appeared once the fencer big toe of the back foot. To determine the sequence began his/her movement and then had time to of the segmentary participation, the time of lifting change the trajectory, before the “point of no the front foot was measured (time of lifting the front return.” Once thought has passed this point, there foot) and the maximum horizontal velocities of the is no turning back; the action is inevitable and it is markers situated on the heel of the front foot and at not possible to measure reaction time (Osman, the beginning of the forte section (lower third) of the Kornblum, & Meyer, 1986). blade (VX(MAX) foot and VX(MAX) epee, respec- To avoid the factor of learning, 9 trials were car- tively), as well as the timing of these velocities, ried out under different conditions: 5 valid trials expressed as percentages with respect to the accel- where the stimulus was changed at a pre-established eration time (Time VX(Max) foot and Time VX(Max) instant (time-S2), 2 trials where no change was epee). Finally, the length of the long lunge was made, and 2 trials where the stimulus was changed recorded under the two experimental conditions at a random time between the reaction time (RT) defined as the distance between the marker of the and time-S2. The order of the trials was random and toe of the back foot in the on-guard position and the a trial was repeated when the target was not reached, marker situated on the heel of the front foot fully misses being noted as errors. Only the 5 valid trials planted on the floor after a lunge. were recorded, in which the target change was made From the force-platform data, the records for velo- at time-S2, and only the trial with the median reac- cities and movements of centre of mass (CM) were tion response time (RRT) of the 5 was analysed. determined following the method of Gutiérrez- Dávila et al. (2006). The horizontal acceleration (ax) of the CM was calculated from the net horizon- Data analysis tal force of the two platforms (FAX and FBX) and the For an evaluation of the results when the target was mass of the fencer. Progressive horizontal velocities changed, in comparison to the simple attack with a (VCM(X)) and displacements (SCM(X)) were calcu- direct thrust, the reaction response time was lated from the horizontal acceleration-time values recorded, as were its two most significant compo- using trapezoidal integration. For the calculation of nents: a) reaction time (RT), defined as the period the horizontal velocity of the markers, the horizontal Downloaded by [UGR-BTCA Gral Universitaria] at 13:15 14 May 2013 from the appearance of the stimulus (S1) until move- component was used after having been filtered by ment begins; and b) the movement time (MT), fifth-order spline functions. The smoothing factor defined as the period from beginning the movement was calculated by the method known as cross- until the instant that the point of the epee touches generalised validation (Woltring, 1985). Then, the the plastron. Choice reaction time (CRT), was first derivative was applied with respect to time, defined as the time period between S2 and the using the fifth-order spline function without applying instant when the fencer changes (Target change) the the smoothing function. transverse or vertical component of the acceleration of the marker situated at the second third of the Statistical analyses epee. For the calculation of the acceleration components, the second derivate of the movement vector For the statistical treatment of the data, the software was used through fifth-grade splines without apply- used was SPSS v. 20.0 software for Windows (SPSS, ing the smoothing function (Figure 1). For the tem- Inc., Chicago, IL). The mean and standard deviation poral analysis, the time needed by the fencer during were calculated for the variables in each experimental horizontal acceleration (time-acceleration phase) condition. To calculate the differences between means and the time during which the fencer is in flight of the variables in the two types of simple attack (sim- (time-flight phase) were recorded. For the analysis ple thrust and target change), a multifactorial analysis 1104 M. Gutiérrez-Dávila et al.

of variance for repeated measures (ANOVA) was used. the target change, the peak of the force is lower, Mean differences between experimental conditions whereas the time needed for the acceleration is and 95% Confidence Interval (CI) were calculated. greater than that needed for the straight thrust. The Effect size statistics was assessed using Cohen’s d, reaction time (RT) is shorter when the attack is taking into account the cut-off established by Cohen, made with the straight thrust (Figure 2(b)). 0.20, 0.50, and 0.80, constitute small, medium, and Table I lists the numerical data corresponding to large effect sizes, respectively (Nakagawa & Cuthill, the temporal parameters of the RT and certain 2007). kinetic factors related to the centre of mass (CM). To evaluate the reliability of the trials, an analysis It was confirmed that the mean RT and the move- of variance for repeated measures was applied to ment time (MT) were significantly less when the all the trials under the two experimental conditions attack was made with the straight thrust (5 trials), taking as the dependent variable the reac- (P < 0.001). As a result, the mean reaction response tion response time (RRT). No significant differences time (RRT) was also significantly shorter for the were found between the two tests. The intraclass straight thrust. Both VX(CM) and SX(CM) decreased correlation coefficient was 0.934 (P < 0.001) for significantly during the first 0.2 s of the acceleration the simple attack with a straight thrust and 0.909 phase when the attack was realised with a target (P <0.001) for the attack with a target change. A change (P < 0.001), which indicates that the hori- P-value of P < 0.05 was considered to correspond zontal velocity of the CM, at the beginning of the to statistical significance. phase of acceleration, is lessened when the simple attack is realised with a target change conditioned to represent the defence of an opponent. The mean Results horizontal velocity of the CM at the end of the Figure 2 presents a typical example of the horizontal- acceleration phase (vCM(X)) was significantly greater force values for one of the fencers during the accel- (P < 0.001) when the attack was a straight thrust eration phase. The upper part of the figure (a) shows and the mean displacement of the CM (sCM(X)) separately the horizontal forces recorded in the two presented a certain significance among the means platforms, while in the lower part (b) appear the (P < 0.05), being greater when the attack was a results of the horizontal forces exerted by the two straight thrust. With respect to the phases in which supports under the two experimental conditions. the movement was divided (acceleration impulse and The movement starts with the increase in force of flight) significant differences were only seen in the the back foot and the reduction of the force of the mean time of acceleration impulse (P <0.001), while front foot. When the simple attack is executed with no differences were found in flight time. Table I also presents the errors made under the two conditions, expressed in percentages of all the trials carried out. The results indicate that when the attack was a direct thrust, the errors significantly diminished (P < 0.001). Table II contains the data of the markers on the hip, shoulder, and the beginning of the forte of the epee blade (near the hilt), with respect to the big toe Downloaded by [UGR-BTCA Gral Universitaria] at 13:15 14 May 2013 of the back foot in the on-guard position. None of the positions registered statically significant differences between the means of the two conditions. Consequently, the mean starting position was similar in both cases. Below, data are presented to describe the characteristics of the movement pattern utilised in both conditions. The time of lifting the front foot was significantly greater when the attack was made with a target change (P < 0.001). No statistically significant differences were found for the times at which the front foot and epee reached maximum velocity (Time VX(MAX) foot and Time VX(MAX) epee, respectively) and therefore, in the two conditions, the foot attained its maximum velocity before Figure 2. Horizontal forces for one of the fencers during the acceleration phase. In the upper part (a) appear the horizontal the epee did. Finally, we should indicate that the forces recorded in the two platforms and in the lower part (b), maximum velocity of the epee was significantly the resulting horizontal forces. greater when the attack was direct (P < 0.001). Effect of target change in fencing 1105

Table I. Descriptive and inferential statistics of the temporal parameters of the response and reaction as well as other relevant variables for the attacks made with direct thrust or target change.

Direct-thrust attack Target-change attack Mean differences Variables N = 30 Mean ± s N = 30 Mean ± s Mean ± s 95% CI Effect size d

Reaction Resp. Time, RRT (ms) 710 ± 82*** 807 ± 81*** −97 ± 60 −119 to −74 1.6 Reaction Time, RT (ms) 188 ± 22*** 216 ± 28*** −28 ± 32 −40 to −16 0.9 Movement Time, MT (ms) 523 ± 74*** 591 ± 76*** −69 ± 50 −87 to −50 1.4 Choice React. Time, CRT (ms) − 234 ± 36 Time-acceleration Phase (ms) 505 ± 68*** 548 ± 72*** −43 ± 49 −62 to −25 0.9 Time-ﬂight Phase (ms) 28 ± 29 21 ± 33 7 ± 20 −0.9 to 14 0.3 Errors (%) 20 ± 17*** 38 ± 15*** −18 ± 19 −25 to −11 0.9

VX(CM) ﬁrst 0.2 s of acceleration 0.52 ± 0.11*** 0.36 ± 0.13*** 0.16 ± 0.14 0.11 to 0.21 1.1 phase (m · s−1)

SX(CM) ﬁrst 0.2 s of acceleration 0.04 ± 0.01*** 0.02 ± 0.01*** 0.01 ± 0.01 0.01 to 0.02 1 phase (m)

VX(CM) end of the acceleration 1.72 ± 0.36* 1.60 ± 0.38* 0.11 ± 0.20 0.03 to 0.19 0.5 phase (m · s−1)

SX(CM) end of the acceleration 0.39 ± 0.08* 0.35 ± 0.09* 0.04 ± 0.08 0.01 to 0.07 0.5 phase (m)

***P < 0.001; *P < 0.05.

Table II. Descriptive and inferential statistics of the main kinematic variables for the situations of direct attack and target-change attack.

Direct-thrust attack Target-change attack Mean differences Variables N = 30 Mean ± s N = 30 Mean ± s Mean ± s 95% CI Effect size d

On-guard position Horizontal position hip (m) 0.38 ± 0.06 0.36 ± 0.07 0.01 ± 0.05 −0.01 to 0.03 0.3 Horizontal pos. shoulder (m) 0.49 ± 0.08 0.49 ± 0.09 0 ± 0.07 −0.02 to 0.02 0 Horizontal position epee (m) 1.06 ± 0.10 1.04 ± 0.12 0.02 ± 0.08 −0.01 to 0.05 0.2 Movement time (MT) Time lifting front foot (s) 0.039 ± 0.001*** 0.083 ± 0.023*** −0.04 ± 0.02 −0.05 to –0.03 2

Time VX(MAX) foot (%) 68 ± 7 66 ± 10 2 ± 9 −50 to 1 0.3 −1 VX(MAX) foot (m · s ) 4.10 ± 0.80 3.98 ± 0.71 0.12 ± 0.62 −0.1 to 0.36 0.2 Time maximum velocity 84 ± 6 87 ± 9 −3±9 −6 to 1 0.3

epee, VX(MAX) epee (%) Maximum velocity 2.49 ± 0.40*** 2.15 ± 0.56*** 0.33 ± 0.35 0.20 to 0.46 1 −1 epee, VX(MAX) epee (m · s ) Lunge length (m) 1.37 ± 0.19 1.34 ± 0.19 0.03 ± 0.35 −0.10 to 0.16 0.1

***P < 0.001.

Downloaded by [UGR-BTCA Gral Universitaria] at 13:15 14 May 2013 Discussion and conclusions In both experimental situations, the lunge length was similar; the weapon reaches maximum velocity The data show that when a target change is made after the foot has attained its maximum velocity. during a simple attack in fencing, the movement This data suggests that the movement pattern was time (MT) increases. The velocity (Vx) and the similar for the two experimental conditions, corro- horizontal distance (Sx) travelled by the CM at the borating the findings of Williams and Walmsley end of the acceleration reduces, while the time (2000), who suggested that the target change did needed for acceleration impulse increases (see not pose overwhelming technical difficulties. Table I). Consequently, when the target changes The time of the phases increased in the target- due to an action of the opponent, the displacement change attack except for the time-flight phase is slower than when the attack is a straight thrust. because the uncertainty led to the fencer putting This aspect is confirmed by the results of Sanderson his/her foot on the floor faster and reducing the (1983) and Borysiuk and Waskiewicz (2008), which velocity of movement. If the time of the phases are show that uncertainty caused by target change dur- analysed in percentage (%) of the reaction response ing the attacking action in fencing reduces velocity. time (RRT), the phases diminish in a similar percen- The graph in Figure 2 for one of the participants tage (Table II), what is considered an invariant of approximates a typical example. the movement (Schmidt & Lee, 2011), with the 1106 M. Gutiérrez-Dávila et al.

exception of the flight phase. Taking these consid- when an attack is executed with no expectation of a erations into account, the velocity of the movement response from the opponent (direct thrust) the is a variant of the same pattern. motor planes and their sequencing arise by a pattern The high level of variability in flight time phase is of only benefit, known as automatic facilitation (Gao justified by the initial position of the fencer in the et al., 2009). However, when there is an expectation experimental set up. Initially, all the fencers placed of a response from the opponent (target change), the themselves on the platform at a distance which was use of automatic facilitation could become a liability 1.5 times their height. After that, with the same pro- more than an advantage and lead to errors. To guar- cedure as Williams and Walmsley (2000), the plas- antee the adaptability of the actions involved in tar- tron was slightly moved so that the fencer could feel get change, the fencer develops another type of comfortable and could proceed with the lunge follow- facilitation, known as strategic or high-order facilita- ing the technique normally used. The fencers reduced tion, which enables the fencer to inhibit the first the distance to the plastron by 0.013 ± 0.2 m. The action of attack to refine the precision of the distance chosen by each participant was maintained response during choice reaction time (CRT). for all the trials in the two experimental situations According to the above explanation and the pre- (direct-thrust attack and target change). cepts of Schluter et al., (1998) as well as Duque The results presented can also be explained using et al., (2010), the anticipation of acting by an inhibi- the dual-visual system proposed by Milner and tion process to resolve conflicts would account for Goodale (1995). Thus, when an attack is made the slower movement during a target change and, with a direct thrust, the dorsal current would be especially, the delay encountered in reaction time. the dominant one, from the start of the movement The findings of Praamstra, Kourtis, Fei Kwok, and (S1). Being a current that takes visual information Ooterveld (2006) support this contention by demon- implicit in the movement, it enables the movement strating that in multiple-choice tasks with a reaction to be executed swiftly and automatically while main- time, participants tend to anticipate implicitly or taining good precision of the motion until reaching unconsciously the time intervals, delaying the reac- the target (Goodale & Westwood, 2004; Van der tion time. Kamp et al., 2008). On the contrary, when the attack In conclusion, when a target change results from is made with the target change, the ventral current, or may prevent the defensive action of the opponent, associated with explicit awareness, dominates over we find that the reaction response time increases, the the dorsal until after the information of the stimulus execution velocity diminishes, and errors increase. change (S2) has been processed. The result is that However, we did not detect changes in the temporal the movement starts more slowly, increasing in velo- sequence of segmentary participation. city only during the MT target change, when the According to the practical applications of the present dorsal current already assumes dominance over the study, the movement time (MT) that a fencer takes for ventral. Although we cannot confirm that this is the a direct trust would be greater than the reaction time cause of the increase in the errors when a target that the opponent would need to respond to this change is undertaken (see Table I), the lack of syn- attack. It appears improbable that with the distance to chronisation and the possible interference in the the target established in this study the attack with a contribution of the two currents (Schneider & straight thrust would be successful. Therefore, these Deubel, 2002), together with the fact of having less types of actions should be trained for with closer objec- Downloaded by [UGR-BTCA Gral Universitaria] at 13:15 14 May 2013 time to make the precise adjustments that would tives (the arm or leg of the opponent), or this attack enable the target to be reached and the small dia- should be preceded with a certain tactical intent to meter of the circle (target) 0.09 m, could be respon- delay the opponent’s response. sible for increased error. The high level of variability According to the above, considering that the target in each experimental condition was due to the varia- change results in a reduction in velocity of the CM bility of the distance to the plastron (target); each and an increase in MT (591 ± 76 ms), the simple fencer felt comfortable and was able to proceed with attack with a target change does not appear to be a the lunge following the technique normally used. basic attack option, but rather would constitute a Despite the above, the fact that the reaction time resource that is determined by uncertainty, the was significantly greater when the attack involved a choice reaction time (CRT), and the time needed target change is difficult to explain using theories to execute the target change (MT target change). related to visual information processing. It bears Therefore, the practical utility for training for high- pointing out that in both conditions the fencers had level competition should be understood as an ele- to respond initially to the same stimulus (S1). The ment within the general attack strategy, with the difference that arises between the two situations was aim of obtaining from the opponent a predictable the complexity of the task after the initial stimulus. erroneous response towards the first objective (S1). From cognitive neuropsychology, it is known that The fact of knowing beforehand the possible Effect of target change in fencing 1107

response of the opponent constitutes a tactical Kreighbaum, E., & Barthels, K. M. (1996). Biomechanics. A qua- element of special relevancy in fencing, that allows litative approach for studying human movement (4th ed.). Boston, a reduction in uncertainty and this is related with the MA: Allyn & Bacon. Milner A. D., & Goodale, M. A. (1995). The visual brain in action. ability to switch, a reaction with a movement pattern Oxford: Oxford University Press. acquired with similar characteristics to a direct thrust Nakagawa, S., & Cuthill, I. C. (2007). Effect size, confidence situation, despite the realisation of a target change. interval and statistical significance: A practical guide for biolo- gists. Biological Reviews Cambridge Philosophical Society, 82(4), 591–605. Acknowledgements Osman, A., Kornblum, S., & Meyer, D. E. (1986). The point of no return in choice reaction time: Controlled and ballistic The authors wish to thank Cristina López de stages of response preparation. Journal of Experimental Subijana for her invaluable help collecting data and Psychology: Human Perception and performance, 12(3), 243–258. the Spanish Royal Federation of Fencing for its col- Praamstra, P., Kourtis, D., Fei Kwok, H., & Ooterveld, R. laboration in this project, providing access to the (2006). Neurophysiology of implicit timing in serial choice fencers of the National Epee fighting Team. reaction-time performance. Journal of Neuroscience, 26(20), 5448–5455. Sanderson, F. H. (1983). The effect of directional uncertainty on References reaction time and movement time in a fencing task. Journal of Sports Sciences, 1(2), 105–110. Borysiuk, Z., & Waskiewicz, Z. (2008). Information processes, Schluter, N. D., Rushworth, M. F., Passingham, R. E., & Mills, stimulation and perceptual training in fencing. Journal of K. R. (1998). Temporary interference in human lateral premo- Human Kinetics, 19, 63–82. tor cortex suggests dominance for the selection of movements. Czajkowski, Z. (2005). Understanding fencing: The unity of theory A study using transcranial magnetic stimulation. Brain, 121, and practice. Staten Island, NY: SKA Swordplay Books. 785–799. Desmurget, M., & Sirigu, A. (2009). A parietal-premotor network Schmidt, R. A., & Lee, T. D. (2011). Motor control and learning: for movement intention and motor awareness. Trends in A behavioral emphasis (5th ed.). Champain, IL: Human Cognitive Sciences, 13(10), 411–419. Kinetics. Di Russo, F., Taddei, F., Apnile, T., & Spinelli, D. (2006). Neural Schneider, W. X., & Deubel, H. (2002). Selection-for-perception correlates of fast stimulus discrimination and response selection and selection-for-spatial-motor action are coupled by visual in top-level fencers. Neuroscience Letters, 408, 113–118. attention. A review of recent finding and new evidence from Duque, J., Lew, D., Mazzocchio, R., Oliver, E., & Ivry, R. B. stimulus driven saccade control. In W. Prinz & B. Hommel (2010). Evidence for two concurrent inhibitory mechanisms (Eds.), Attention and performance XIX: Common mechanisms in during response preparation. Journal of Neuroscience, 30(10), perception and action (pp. 609–627). Oxford: Oxford University 3796–3802. Press. Gao, J., Wong-Lin, K., Holmes, P., Simen, P., & Cohen, J. D. Stewart, S. L., & Kopetka, B. (2005). The kinematic determinants (2009). Sequential effects in two-choice reaction time tasks: of speed in the fencing lunge. Journal of Sports Sciences, 23(2), Decomposition and synthesis of mechanisms. Neural 105. Computation, 21(9), 2407–2436. Thirioux, M. P. (1970). Escrime moderne aux trois armes. Modern Goodale, M. A., & Westwood, D. A. (2004). An evolving view of fencing to three weapons Paris: Amphora. duplex vision: Separate but interacting cortical pathways for Van der Kamp, J., Rivas, F., Van Doorn, H., & Savelsbergh, G. perception and action. Current Opinion in Neurobiology, 14, (2008). Ventral and dorsal system contributions to visual 203–211. anticipation in fast ball sports. International Journal of Sport Gutiérrez-Dávila, M., Dapena, J., & Campos, J. (2006). The effect Psychology, 39, 100–130. of muscular pre-tensing on the sprint start. Journal of Applied Williams, L. R. T., & Walmsley, A. (2000). Response amendment Biomechanics, 22(3), 194–201. in fencing: Differences between elite and novice subjects. Ivanoff, J., Branning, P., & Marois, R. (2009). Mapping the path- Perceptual and Motor Skills, 91(1), 131–142. ways of information processing from sensation to action in four Woltring, H. J. (1985). On optimal smoothing and derivate esti- Downloaded by [UGR-BTCA Gral Universitaria] at 13:15 14 May 2013 distinct sensorimotor tasks. Human Brain Mapping, 30(12), mation from noisy displacement data in biomechanics. Human 4167–4186. Movement Science, 4, 229–245.